Psychiatric and psychosocial problems in adults with normal-intelligence autism spectrum disorders

What happens to someone with autism when they grow up? Not what childhood looks like — that story has been told many times. Instead, we focus on the adult years: the psychiatric weight, the daily life, and the relationships. For a long time, almost no one had looked at this. Most autism research focused on children. Adult studies, when they existed at all, concentrated on people with intellectual disabilities. The slice of the population that had normal intelligence and simply aged out of pediatric attention was, as Hofvander and colleagues put it, largely absent from the literature. Reported estimates of co-occurring psychiatric disorders in this group had ranged anywhere from 9 to 89 percent. That spread isn't a sign of a rich literature; it's a sign of a thin one. Hofvander and colleagues set out to close that gap with a carefully built clinical sample: 122 consecutively referred adults, all meeting DSM-IV criteria for an autism spectrum disorder, and all with normal intelligence. The group broke down into five patients with autistic disorder, 67 with Asperger's disorder, and 50 with pervasive developmental disorder not otherwise specified, known as PDD-NOS, which is the catch-all for presentations that don't fully meet the other two sets of criteria. Eighty-two of the 122 were men, 40 were women, and the median age was 29, with a range stretched from 16 to 60. Patients were drawn from two specialist centers, one in Gothenburg and one in Paris, and the protocols were similar enough that pooling them was judged reasonable. The diagnostic approach was rigorous and deliberately doubled. Every subject was assessed against both DSM-IV criteria and the Gillberg and Gillberg research criteria — a parallel system that reproduces the core DSM triad but also flags verbal peculiarities and abnormal motor development. The Asperger Syndrome Diagnostic Interview was used in 87 percent of subjects. Global intelligence was measured in 93 percent of the sample using the Wechsler Adult Intelligence Scale Revised or the Wechsler Adult Intelligence Scale Third Edition. Structured clinical interviews covered lifetime and current psychiatric diagnoses, and personality disorders were assessed in 96 percent of patients. The point of all this layering was to produce a picture detailed enough to actually be useful. What it revealed was striking on nearly every front. Let’s start with the core symptoms, because the first major finding is about persistence. The defining features of autism — impaired social interaction, deficits in communication, and restricted and repetitive behaviors — did not fade with age. Across all three diagnostic subgroups, 98 percent of subjects met the social interaction criterion. Non-verbal communication problems were present in 89 percent. Restricted and repetitive behaviors were met by 81 percent under DSM-IV and 84 percent under the Gillberg and Gillberg narrow-interests criterion. These aren't subtle residues. They reflect the full clinical picture, still present in adults in their thirties, forties, and fifties. The most counterintuitive finding here involves Asperger's disorder specifically. By DSM-IV definition, Asperger's is distinguished from autistic disorder partly by normal early verbal development. Language isn't supposed to be a major problem. Yet, 51 percent of the Asperger's subjects in this study met the DSM-IV communication criterion anyway. Using the Gillberg and Gillberg formulation, that figure climbed to 84 percent. Verbal and speech-language atypicalities were common in a group that, by diagnostic definition, wasn't supposed to have them. That's not a minor wrinkle; it raises real questions about whether the subtype boundaries drawn in DSM-IV actually carve the population at its natural joints. The authors say as much, arguing the data fit a dimensional model better than a categorical one. The comorbidity findings are where the picture becomes clinically urgent. Hofvander and colleagues found that lifetime psychiatric comorbidity was not occasional or incidental; it was the norm. Fifty-three percent of the sample — 65 people — met lifetime criteria for a mood disorder. Fifty percent met criteria for at least one anxiety disorder. Attention-deficit/hyperactivity disorder was present in 43 percent of the group. Psychotic disorders, which might surprise people given the distinct nature of those conditions, appeared in 12 percent — that's 15 individuals with diagnoses including schizophreniform disorder and brief psychotic disorder. Furthermore, 62 percent of those assessed for personality disorders met criteria for at least one. What makes those numbers particularly important is their consistency. The frequency of mood disorders, anxiety disorders, ADHD, and psychotic disorders did not differ significantly between the three autism spectrum disorder subgroups. It did not differ significantly between men and women either. Whatever is driving this psychiatric burden appears to cut across the diagnostic distinctions that clinicians use and across sex. There were two notable exceptions to this uniformity. Substance use disorders were significantly more common in the PDD-NOS group than in the Asperger's group. Antisocial personality disorder was also elevated in PDD-NOS. But the headline finding is the consistency; the broad psychiatric load was shared across the spectrum. One practical detail from the treatment history data underlines how underserved this population has been. Only 34 percent of subjects had ever received antidepressant treatment, despite 53 percent meeting lifetime criteria for a mood disorder. Just 2 percent had ever received a mood stabilizer. Then there is the psychosocial picture, where the gap between cognitive ability and lived outcome becomes impossible to ignore. Two-thirds of the sample had completed upper secondary school. Twenty-four percent had finished college or university. These are not people who lacked the intellectual capacity to navigate adult life. Yet, at the time of assessment, only 43 percent were employed or studying. More than half were unemployed, on sick leave, or receiving a medical pension. Around 40 percent were still living with their parents or in community-based group homes. Romantic relationships were rare. Only 16 percent — 19 people out of 122 — had ever lived in a long-term relationship. Additionally, a majority, 56 percent of the sample, reported having been bullied at school. That figure was not evenly distributed; female subjects reported school bullying significantly more often than male subjects, with a chi-squared value of 6.09. Intelligence, in this data, does not protect against social exclusion. Education does not reliably translate into employment. And the social deficits that define autism in childhood do not quietly resolve when a person turns 18. The authors are careful about the limits of what they can claim. There was no control group. The two clinical sites had slightly different protocols, and disorder frequencies differed between them. Because the sample was clinically referred, many subjects had prior psychiatric contacts, meaning comorbidity rates could be overestimates of what you would find in the broader population of normal intelligence adults with autism spectrum disorders. These are real constraints. But the core clinical message stands, regardless of whether the exact percentages shift somewhat in a population-based study. The practical implication Hofvander and colleagues push hardest is about diagnostic rules. DSM-IV, at the time of this study, excluded certain concurrent diagnoses in the presence of autism spectrum disorder, ADHD being the most important example. The team deliberately set those exclusion rules aside, and what they found underneath was a clinically real landscape of overlapping conditions. Their argument is direct: those exclusion criteria should be reexamined. If you prohibit the diagnosis of ADHD alongside autism spectrum disorder, you don't make ADHD disappear from that person's life. You just make it invisible in the clinical record. What this study ultimately shows is that autism spectrum disorder in normal intelligence adults is not a single, bounded condition with a tidy comorbidity profile. It is a condition that persists across the lifespan, with its core features largely intact. It carries a heavy and broadly distributed psychiatric burden, and despite the intellectual ability of the people who have it, it produces adult lives characterized by unemployment, dependence, social isolation, and a history of being victimized. The research questions that were mostly being asked about children needed, all along, to be asked about adults too. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.